Post on 19-Jun-2021
SPECTROSCOPIC INVESTIGATION OF PROTEINS
B.S. in Chemistry, LaRoche College, 2004
Benjamin Kabagambe
by
Master of Science in Chemistry
of the requirements for the degree of
College of Arts and Sciences in partial fulfillment
Submitted to the Graduate Faculty of
2007
University of Pittsburgh
FACULTY OF ARTS AND SCIENCES
UNIVERSITY OF PITTSBURGH
Benjamin Kabagambe
by
This thesis was presented
Thesis Advisor: Sanford A. Asher, Distinguished Professor, Department of Chemistry
Adrian Michael, Professor, Department of Chemistry
Stephen Weber, Professor, Department of Chemistry
and approved by
October 12, 2007
It was defended on
ii
2007
Copyright © by Benjamin Kabagambe
iii
SPECTROSCOPIC INVESTIGATION OF PROTEINS
Benjamin Kabagambe, M.S.
University of Pittsburgh, 2007
Apomyoglobin is obtained from pure myoglobin by extracting out the heme group. Using
UV-visible spectrometry we are able to monitor complete removal of heme group. This
extraction initiates the disruption of myoglobin’s tertiary conformation. Apomyoglobin consists
of 8 α-helices labeled A through H. These helices unfold when the protein is subjected to pH
decrease. It is more compact at about pH 7 and unfolds as we change to more acidic
environment. At about pH 2 we have A, G, H core with the rest of the helices unfolded. At pH
values slightly lower than 2, A, G and H helices are also unfolded. When the protein is not
denatured, the refolding process is done by changing the pH towards neutral. At pH 2, G and H
helices refold and at pH 4 A-helix refolds as well. We have been able to label A-helix with
protons and the rest of the protein with deuterium using H2O and D2O respectively. We have
then used temperature change to initiate the unfolding of protonated A-helix. When dissolved in
D2O, the protonated A-helix at elevated temperatures, exchanges its N-H protons to N-D
deuterons. Temperatures were elevated from 0˚ C to 80˚C and changes in Am III peak intensities
were observed using UV resonance Raman spectroscopy. These peak changes were quantified
and used to calculate the number of N-H bonds present between 0˚ C and 80˚C. The number of
N-H bonds decreased with increase in temperature indicating the unfolding of A-helix.
iv
TABLE OF CONTENTS
ACKNOWLEDGEMENTS. ...................................................................................................... XI
1.0 INTRODUCTION TO PROTEIN FOLDING. ......................................................... 1
1.1 ELECTROSTATIC INTERACTIONS .................................................................. 1
1.2 HYDROGEN BONDING ......................................................................................... 2
1.3 HYDROPHOBIC INTERACTIONS ...................................................................... 2
1.4 DISULFIDE BRIDGES ............................................................................................ 3
1.5 VAN DER WAALS INTERACTION ..................................................................... 3
1.6 METHODOLOGY FOR FOLDING STUDIES .................................................... 4
1.6.1 UV RESONANCE RAMAN SPECTROSCOPY ....................................... 5
2.0 RAMAN EFFECT: CLASSICAL AND QUANTUM MECHANICAL
TREATMENT ............................................................................................................................... 7
2.1 INTRODUCTION..................................................................................................... 7
2.2 CLASSICAL TREATMENT ................................................................................... 9
2.3 QUANTUM MECHANICAL TREATMENT ..................................................... 11
2.3.1 INTRODUCTION....................................................................................... 11
2.3.2 TIME DEPENDENT PERTURBATION THEORY AND TRANSITION
POLARIZABILITY fi)(α ......................................................................................... 12
v
2.3.3 TYPES OF RAMAN SCATTERING PROCESSES AND
POLARIZABILITY fi)(α ......................................................................................... 15
3.0 UV-RESONANCE RAMAN STUDY OF THE UNFOLDING BEHAVIOR OF
APOMYOGLOBIN .................................................................................................................... 18
3.1 ABSTRACT ............................................................................................................. 18
3.2 INTRODUCTION................................................................................................... 19
3.3 EXPERIMENTAL .................................................................................................. 20
3.4 RESULTS AND DISCUSSION ............................................................................. 25
3.5 CONCLUSION ....................................................................................................... 31
4.0 FUTURE WORK ....................................................................................................... 32
4.1 TIME RESOLVED MEASUREMENTS OF APOMYOGLOBIN
UNFOLDING ...................................................................................................................... 32
REFERENCES ............................................................................................................................ 34
vi
LIST OF TABLES
Table 1. Calculated charges z obtained from the mass spectrum of apomyoglobin in figure 6. The
formula used for this calculation is shown on the table. The molecular weight (mw) of horse
heart myoglobin in this case is 16947 g. ....................................................................................... 24
Table 2. Table shows Amide I’, and III peak heights obtained from figure 11. These intensities
IAmI’ and IAm III, were obtained by normalizing all spectra with respect to Am I’ band and using a
ruler to measure them as shown in figure 11 above. The last column is obtained using equation
26 (also explained in figure 12). ................................................................................................... 29
vii
LIST OF FIGURES
Figure 1 Illustration of scattering of an incident electromagnetic radiation from a laser source.
Asher S. A. 32 .................................................................................................................................. 8
Figure 2. These are linked electric dipole transitions the product of which is shown in equation
22 numerator. are energy levels of the molecule before and after
perturbation ................................................................................................................................ 13
>>> randfi ||,|
Figure 3. Raman scattering process types as a result of proximity of frequencies riandωω1 .
Adopted from Long D. A. reference 31 ........................................................................................ 15
Figure 4. Model for acid denaturation of apoMb and holoMb. The helical structure is represented
as ribbons. Lines represent unfolded structures or the so-called β-strands. Adapted from Chi Z.
and Asher S.A.34 ........................................................................................................................... 20
Figure 5. Absorption spectra of myoglobin before and after extraction of heme group. The Soret
band at 400 nm disappears which indicates loss of the heme group. Concentration of the original
solution was 5 mg/ml. ................................................................................................................... 21
Figure 6. HPLC was used to purify apomyoglobin. The top spectrum is obtained by using 220
nm detector and bottom uses the 278 nm detector. The peak at 31st minute appears in both
spectra and corresponds to apomb. ............................................................................................... 22
viii
Figure 7. Mass spectrum of Apomyoglobin. The above m/z ratios are indicative of a multi-
charged molecule. The charges on this protein are as high as 29+ as shown in a table that
follows........................................................................................................................................... 23
Figure 8. CD spectra of apoMb at different pH values from 1.74 to 6.30. It shows no helical
conformation at pH 2 and below. We see α-helix conformation appear at pH 4 and above.
Concentrations are about 20 µM ApoMb in D2O. Temperature was 25º C. ................................ 25
Figure 9. UVRRS excited at 204 nm at 10º C: (A) ApoMb after 16hrs in D2O, pD 4.0.
(B)ApoMb in H2O, pH 4.0. (C) Difference spectra. .................................................................... 26
Figure 10. 204 nm UV resonance Raman spectra of apoMb sample measured between 0 and 80
ºC. Three 5 min spectra were recorded and summed. They were normalized with respect to the
Am I’ band. The change in AmIII region is due to peptide bond deuteration which is facilitated
by A-helix melting. ....................................................................................................................... 27
Figure 11. Raman spectra used to estimate peak intensities IAmI’ and IAmIII in the Am I’ and Am
III regions, respectively. Am I’ peaks are used to normalize the spectra. The peak intensity
results are shown in table 2 below. ............................................................................................... 28
Figure 12. This graph is plotted using data from table 2 which were generated from equation 26.
IAmI’ and IAmIII were measured in figure 11 and shown in table 2. The number of N-H bonds
contributing to Am III band is nAmIII and to Am I’ band is nAmI’, =152 peptide bonds/protein. The
α-helix cross sections are σ AmI’ =32 mbarn molecule-1 sr-1 and σ AmIII =24 mbarn molecule-1 sr-1,
obtained from Chi et al.(39). Also superimposed is the dependence on temperature of Am III
intensity normalized with respect to Am I’. .................................................................................. 30
ix
Figure 13. T-jump spectrometer consisting of an Nd: YAG laser, 2 H2 Raman shifters (R1 and
R2), a thermostated flow cell sample circulator (FC) and an ICCD detector (SP).reproduced from
Lednev I. K. reference38 ................................................................................................................ 32
x
xi
ACKNOWLEDGEMENTS.
I would like to express gratitude to my thesis advisor, Professor Sanford A. Asher, for the
support, knowledge, and encouragement that he generously provided during my graduate studies
at the University of Pittsburgh. Thanks to NIH for funding my project.
I would also like to acknowledge Zeeshan Ahmed for giving me his scientific insight and
encouragement. I like to thank Alex Mikhonin for great discussions and ideas he gave me on my
project. Special thanks to all past and present members of Asher research group for they
contributed to my growth as a scientist.
I gratefully acknowledge all professors who walked me through all graduate course work.
Special thanks to the department of chemistry for giving me a chance to teach undergraduate
students. It was one of my greatest learning experiences.
I would also like to recognize my undergraduate advisors Dr. Roberta Hartman and Dr.
Don Fugito for showing me the right path. They both advised me to pursue a chemistry career.
Special thanks to Mike Sandala for rewarding my work with a research chemist position at PPG
Industries.
Finally, I would like to extend my sincere gratitude to my wife Dorothy Mutoni and our
son Ian Kabagambe for the love and patience they gave me during my graduate studies. Thanks
to my mother Vivian Kayitesi, my sisters Sarah Nakazana, Babra Mugisha, and Resty Kayitesi
for all the support.
1.0 INTRODUCTION TO PROTEIN FOLDING.
Protein folding is one of the most fundamental problems of the 21st century. Studies of
protein folding provide a better understanding of molecular processes underlying disease which
in turn helps in better drug design. These studies are complicated by the fact that the system
presents innumerable degrees of freedom. The interaction of the peptide geometry determines the
conformation of the peptide. The thermodynamic factors that drive high entropy (unfolded state)
or low entropy (native state) of the system have to be understood in order to better understand
the folding process. In the following sections we describe some of these factors and how they
affect the folding process.
1.1 ELECTROSTATIC INTERACTIONS
Electrostatic interactions play a significant role in protein folding. These are interactions
between ionizable side chains e.g. carboxylate and ammonium ions on aspartic and lysine side
chains, respectively. These interactions form ionic bonds-salt bridges. However, most charged
amino acids lie on the protein surface and it’s rare to find interior ionic bonding. Although rare,
the ionic bonds can be important to protein structure especially in peptide structures where the
positively charged N and the negatively charged O are a few angstroms away from each other.
1
1.2 HYDROGEN BONDING
Hydrogen bonds result from interactions between atoms bearing partial negative charge
and partially positively charged hydrogen atoms. The 3-D structures of proteins are heavily
dependent on a variety of networks of H-bonds. The H-bond interactions could be between
atoms on two different amino acid sidechains. It could also be between amino acid sidechain and
solvent molecules at the protein surface. Studies have shown that the native state secondary
structure is a result of hydrogen bonds between backbone carbonyl (C=O) and amide (N-H).1
These intramolecular hydrogen bonds are also responsible for formation and stabilization of α-
helix. Furthermore, hydrogen bonding between two adjacent β-strands leads to β-sheet structure.
1.3 HYDROPHOBIC INTERACTIONS
One major force driving proper protein folding is hydrophobic interactions. These
interactions form an interior hydrophobic core in which non polar sidechains closely associate
and minimize their interaction with polar solvent molecules. All non polar chains are found
inside the backbone where they are not exposed to the polar solvent. When they come to the
surface they are usually involved in other extensive hydrophobic interactions.
The concept behind hydrophobic collapse is explained by Frank and Evans frozen
“Iceberg model.”2-4 When dissolved in water, the non-soluble solutes are surrounded by water
molecules which gain an ordered form and are considered to have low entropy. In hydrophobic
collapse the hydrophobic parts of the protein come together expelling the water molecules with
more entropy. As a result the system gains free energy which is referred to as liberation energy
2
thus, making the collapse a spontaneous process. The collapse results in isolation of non-polar
solutes thus, resulting in a compact state which has less degrees of freedom. This enhances fast
folding.
1.4 DISULFIDE BRIDGES
Disulfide bridges are usually formed by two sulfylhydryl (-SH) groups of cysteine
residues. They are usually found in tertiary globules where their purpose is basically to provide
extra stability to the tertiary structure.
1.5 VAN DER WAALS INTERACTION
The Van der Waals interactions are the weakest of all interactions mentioned above. They
result from the existence of an electron cloud around the atom. The shifts of the electron cloud
around the nucleus yields a temporary electric dipole on an atom. The transient dipole in one
atom induces a complementary dipole in another atom, provided the two are very close to each
other. These forces are weak but their large number makes them an important player in
determining protein structures. Most atoms in proteins are packed very close to each other and
therefore they are involved in forming these Van der Waals interactions. These Van der Waals
interactions along with H-bonding and others are biologically recognized in the antibody-antigen
recognition where the “lock and key” fit of two molecules.
3
1.6 METHODOLOGY FOR FOLDING STUDIES
The driving forces behind protein folding have led researchers to develop all kinds of
models to help them have a better understanding of the process.5-12 The “New View”1 model for
example, holds that, folding is an ensemble in which the peptide chain has large conformational
entropy at the beginning of folding and then hydrophobic collapse and folding process leads to a
spectacular decrease in conformational space for the system facilitating fast folding.1 Folding
continues in a funnel-like energy landscape until the native state is achieved.
In recent years many different experimental techniques have been developed to directly
probe the protein energy landscape. The most prominent techniques in place are X-ray
crystallography13, NMR 14, Fluorescence15, Infrared Spectroscopy (IR)16, Circular Dichroism17,
Mass Spectrometry18, and UV Resonance Raman spectroscopy (UVRRS).
The most detailed information about three dimension structures of proteins is given by X-
ray crystallography, but it requires well defined crystal structures, however, the crystal structure
may not provide information relevant to human needs. Furthermore, it prevents the study of the
dynamic behavior of proteins. In theory NMR can provide detailed information on local and
global protein structure, however, the technique suffers from spectral congestion and
broadening.19 In order to perform NMR experiments, there is need for high peptide/protein
concentrations and this can induce aggregation in large proteins like myoglobin. In addition there
is a time scale that limits kinetic resolutions to milliseconds time regime.1 These problems are
usually overcome by using expensive and laborious isotopic labeling and combined
multidimensional pulsed NMR spectrometers.
4
Circular dichroism provides information on conformation of secondary structure,
however, there is strong interference from transition of aromatic residues which can
confound quantification of secondary structure content.20, 21
∗→ ππ
UVRRS offers the best novel approach of determining equilibrium conformation and
peptide folding dynamics. It gives better information content, and provides relative ease of
handling and sample preparation as will be seen in the next sub-section. It also has the potential
to examine experimental time scales from picoseconds to hours.22
1.6.1 UV RESONANCE RAMAN SPECTROSCOPY
Ultraviolet resonance Raman (UVRR) spectroscopy has typically been used to study
small isolated secondary structures providing detailed information and basic steps required for
studying biomolecular structure and function.23, 24 Chi et al. developed a procedure for
determining the percentage of α-helix and β-sheet composition in dilute solutions. They used the
206.5 nm light to excite within the region and acquired resonance enhanced spectra.25
There are amide bands in resonance enhanced spectra that provide a convenient way of
characterizing the protein backbone conformation. These bands include amide I, II, III and CαH
vibrations which result from secondary structure. These bands are used to characterize the
protein structure and dynamics, for example an increase in CαH band intensity is an indicator of
unfolding.
∗→ ππ
UVRRS based T-jump spectrometer is used to perform time-resolved measurements.
Lednev et al.26 used this spectrometer to do kinetic measurements of AP peptides with temporal
resolution of about 3 ns and were able to see the early steps of protein unfolding. Asher et al.
have recently followed the AP peptide melting profile and found that it melts from a helical
5
structure to a PPII helix conformation.1 This paper has offered support to the recent theoretical
studies that have suggested that small peptides in their unfolded state27, 28 have more PPII helix
conformation.29, 30
In later chapters of this document, UVRRS is used to study the unfolding of
apomyoglobin. Using 204 nm light, we excited the amide regions of the protein. We then
decoupled some motions in the amide backbone by changing the protein environment from H2O
and D2O as illustrated in chapter 3. UVRRS methodology provides hope to improve our
understanding of protein folding in the future.
6
2.0 RAMAN EFFECT: CLASSICAL AND QUANTUM MECHANICAL
TREATMENT
2.1 INTRODUCTION
The Raman Effect is a light scattering phenomenon in which the incident and scattered
light possess different frequencies as seen in figure 1. Unlike in absorption processes, in
scattering processes, the energy of the incident photon doesn’t have to be equal to the difference
between two discrete energy levels of the system. Experiments have shown however, that as the
energy of the incident photon gets closer to an electronic transition energy associated with the
transition from ground to excited state of a system, the scattering intensity is enhanced.31 This
enhancement increases rapidly as the incident photon energy approaches an electronic transition
energy and this is what is termed resonance scattering. Resonance scattering differs significantly
from normal scattering as will be discussed in detail in this chapter.
7
Figure 1 Illustration of scattering of an incident electromagnetic radiation from a laser source.
Asher S. A. 32
In both the classical and quantum mechanical treatments, the scattered radiation
originates from electric and magnetic multipole moments induced in a molecule by
electromagnetic fields of the incident radiation. The most significant contribution is from an
oscillating electric dipole. The dipole moment induced in a molecule by an electric field of
incident radiation of frequency 1ω , oriented at an angle θ with the axis of the dipole, gives
intensity I shown in equation 1.
θωω 220
4' sinpkI s= (1)
300
2'
321
ck
επω = (2)
The term is the amplitude of the induced electric dipole with frequency0p sω . Classical and
quantum mechanical treatments have been done to find how sω and relate to properties of the
scattering molecule and the incident electromagnetic radiation of frequency
0p
1ω .
8
2.2 CLASSICAL TREATMENT
The objective of this section is to calculate the frequency dependent linear induced dipole vector
using its relationship with electric field vector of the incident radiation)1(p E .
Ep ⋅= α)1( (3)
The polarizability tensor α of a molecule is a function of nuclear coordinates. In this treatment
we will assume a space-fixed molecule in its equilibrium configuration with the nuclei vibrating
about the equilibrium position. The polarizability varies with the vibrations of the molecule in
away expressed in the following equation.
...)(21)()(
,0
2
00 lklk lkk
kk
QQQQ
QQ ∑∑ ∂∂
∂+
∂
∂+= ρσρσ
ρσρσ
αααα (4)
Where 0)( ρσα represents the value of ρσα at equilibrium configuration and and are
normal coordinates of the molecular vibrational frequencies
kQ lQ
kω and lω respectively. The above
equation is a Taylor series summation of all normal modes. We will ignore all modes with Q
higher than the first order. We may therefore write this as:
kkk Q)()()( '0 ρσρσρσ ααα += (5)
Where
0' )()(
kk Q∂
∂= ρσ
ρσ
αα (6)
Now assuming mechanical harmonicity, the time dependent coordinate is given by the
equation;
kQ
)cos(0 kkkk tQQ δω += (7)
9
kkQ δ,0
are normal coordinate amplitude and phase factor respectively. To find obtain the time
dependent ρσα resulting from one particular kth molecular vibration we substitute equation 7 into
5:
)cos(0
'0 kkkkk tQ δωααα ++= (8)
At this point we bring in the frequency dependence of electric field E written as
tEE 10 cosω= (9)
And equation 3 becomes
ttQEtEp kkk 10100)1( cos)cos('cos
0ωδωαωα ++= (10)
Using the trigonometric function
)cos()cos(21coscos BABABA −++= (11)
Equation 10 becomes
)cos(2
'cos 1
0100
)1(kk
k ttE
tEp δωωα
ωα ±±+= (12)
This is the same as
)cos(cos 1010)1(
kkRamRay ttptpp δωωω ±±+= (13)
Where
0000 EEp RayRay αα == , and 00 '21 EQp kk
Ram α= (14)
It can be clearly seen from the equation 13 that there are three components that correspond to
Rayleigh, Stokes Raman and anti-stokes Raman scattering. The term gives rise to
radiation at frequency
tp Ray10 cosω
1ω called Rayleigh scattering. The term corresponds tk )cos( 1 ωω +p Ram0
10
to radiation at frequency ( kωω +1 ) and accounts for anti-Stokes Raman scattering. Finally,
corresponds to radiation at frequency (tp kRam )cos( 10 ωω − kωω −1 ) and corresponds to Stokes
Raman scattering. 31
We therefore conclude that classical theory gives a short explanation on scattering of
light of frequency 1ω incident on a molecule. A harmonically oscillating induced dipole moment
results from the interaction between the electric field of light and the molecule. The
fluctuation of the induced dipole moment results in the scattering of light at frequencies
(
)1(p
kωω +1 ) and ( kωω −1
)1(p
fip )( )1(
) which correspond to anti-Stokes and Stokes Raman scattering
respectively. We have also seen from equations 1 and 3 that the polarizability is proportional to
the square root of the scattered light intensity. A more detailed treatment of light scattering is
however done using quantum mechanics as shown in the next section.
2.3 QUANTUM MECHANICAL TREATMENT
2.3.1 INTRODUCTION
While classical theory provides the origin of Raman Effect, quantum theory has a more
detailed view. In this treatment, Raman scattering is described by a time dependent perturbation
theory in which an incident electric field perturbs the molecular eigenstates. The induced electric
dipole moment seen in classical treatment is replaced by a transition electric dipole moment
associated with a transition in a molecule from initial state i to a final state f that have
11
been induced by the incident electric field of frequency 1ω . The total induced transition electric
dipole vector is given by the equation:
...)()()()( )3()2()1( +++= fifififi pppp (15)
Where,
EEEp
EEp
Ep
.61
.21.
)3(
)2(
)1(
γ
β
α
=
=
=
(16)
)3()2()1( ,, ppp are linear, quadratic and cubic in E respectively. We are therefore, able to obtain
expressions for the first-order induced transition electric dipole moment and its
corresponding transition polarizability
fip )( )1(
fi)(α as will be seen in equation 22
2.3.2 TIME DEPENDENT PERTURBATION THEORY AND TRANSITION
POLARIZABILITY fi)(α
The time dependent Schrödinger equation is given as:
=ΨH iħ )(dtdΨ (17)
H is the Hamiltonian operator. is a time-dependent wave function and can be expressed as a
function of the time-independent wave function by the following equation:
Ψ
)exp()0( ti iii ωψ −=Ψ (18)
tWhere iψ is the time-independent wave function and iω is the energy of state i. The perturbed
wave function can be expressed as a linear combination of a series of perturbation terms: 'iΨ
12
...' )2()1()0( +Ψ+Ψ+Ψ=Ψ iiii (19)
The total transition electric dipole is thus given by the following operation.
>ΨΨ=< '||')(^
iffi pp (20)
From equation 15, the first-order induced transition electric dipole moment is thus obtained by:
>ΨΨ<+>ΨΨ=< )1(^
)0()0(^
)1()1( ||||)( ififfi ppp (21)
Where are unperturbed time-dependent wave functions of the initial state i and final f
respectively and are corresponding first-order perturbed time-dependent wave
functions. is the electric dipole moment operator. If we assume the virtual states (r) which
arise after perturbing a molecule, then obtain a more detailed equation 22. The figure below
shows the transition pathway
)0()0( , fi ΨΨ
^p
)1()1( , fi ΨΨ
Figure 2. These are linked electric dipole transitions the product of which is shown in equation
22 numerator. are energy levels of the molecule before and after
perturbation
>>> randfi ||,|
We can expand equation 21 by including the time-independent components using equation 18.
13
=)( )1(p (1/2ħ) tiEi
iprrpfi
iprrpfs
fir rrfrri
ωωωωω σ
ρσσρ −⎪⎭
⎪⎬⎫
⎪⎩
⎪⎨⎧
Γ++>><<
+Γ−−
>><<∑≠
exp||||||||
0, 11
^
+ C.C. (22)
From equation 16 we can see that the polarizability is a function of the electric dipole moment.
=fi)( ρσα (1/ħ) ||||||||
, 11∑≠ Γ−+
>><<+
Γ−−
>><<
fir rrirri iiprrpf
iiprrpf
ωωωωρσσρ (23)
In the above equations, the notation has been changed where stand for || fandi <>
|| fi and ψψ <> respectively. We have also taken virtual states r into consideration. These
virtual states are induced by a perturbation and have lifetimes that relate to shown in the
denominators of the above equation. The lifetime
rΓ
rτ is related to rΓα through the uncertainty
principle. Thus =rτ 1/ ( ). In the initial and final states it is assumed that the lifetimes are
infinite so that .
)(2 ErΓ
0=Γ=Γi f
14
2.3.3 TYPES OF RAMAN SCATTERING PROCESSES AND
POLARIZABILITY fi)(α
Figure 3. Raman scattering process types as a result of proximity of frequencies riandωω1 .
Adopted from Long D. A. reference 31
Four types of scattering processes are described in this section. The polarizability
equation 23 is used to describe the rise of the four different scatters. The denominator of this
equation ( rri iΓ−− 1ωω ) plays a big role in this process. The first instance is where the exciting
radiation 1ω is much smaller than the absorption frequency riω (i.e. riωω <<1 ) for all r. This
makes riri ωωω ≈− 1 for all states r and rΓ can be neglected because they are very small
compared to riω . This is normal Raman scattering shown in fig 3 (a) and is a result of the
molecule interacting with an incident radiation 1ω , making a transition from the initial stationary
15
state to a so-called virtual state r from where it subsequently makes another transition to the final
stationary state f. The virtual state r is not a solution to the time-independent Schrödinger
equation and it doesn’t have a well defined value of energy like stationary states i and f do. The
absorption process that involves no energy conservation but gives rise to such virtual states is
called virtual absorption.
The frequency 1ω approaches a molecular transition frequency riω and results in what is
termed pre-resonance Raman shown in figure 3(b). As riωω ≈1 we get discrete resonance Raman
scattering as shown in figure 3(c). If frequency 1ω is large enough to make the dissociative
continuum energy levels of the system the system is called continuum resonance Raman and is
illustrated in figure 3(d).
The difference between normal and resonance Raman intensities is a result of how large
the excitation frequency 1ω is compared to the molecular transition frequency riω and is
explained by the denominator of the above equation. Resonance Raman intensity is usually
larger than normal Raman intensity because in the first case the denominator becomes very small
when riωω ≈1 (i.e. the denominator is rΓ dependent) and leads to a larger fi)(α whereas in the
second case riωω <<1 the denominator is dependent on riω (i.e. 1ω and are too small and
negligible) hence a smaller
rΓ
fi)(α value. We therefore say that in normal Raman fi)(α is
determined by a weighted sum over states of the products < and the
weighting for each state | is inversely proportional to (
>r| >i>< pr |rf ||^
σ
r
p|^
ρ
i>r ri Γ−− 1ωω ). Thus normal Raman
involves all possible pathways through state that connect the initial state and the final
state as seen in figure 2.
>r| >i|
>f|
16
The intensity expression is generally written as:
(24) )(0 Ω= WNII Rfi σ
NR ,σ and represents the total differential cross section ( , number of
molecules per unit volume and an experimental parameter which is a function of solid collection
angle, (Ω).33 The following equations can be used to calculate the cross section from
intensities obtained from an experiment in which there is a mixture of the sample and an internal
standard.
)(ΩW .)./( 2 srmolccm
Rσ
kcstdIstdkstdcstdI
fi
RfiR
.).()().().(.σ
σ = = cstdI
stdcstdI
fi
Rfi
).()().(.σ
(25)
In the above equation c stands for concentration, std is the internal standard deviation, and k
represents . The cross sections obtained are used to compare the bands from the experimental
spectra as will be seen in the experimental section in later chapters of this pamphlet.
0I
17
3.0 UV-RESONANCE RAMAN STUDY OF THE UNFOLDING BEHAVIOR OF
APOMYOGLOBIN
3.1 ABSTRACT
Apomyoglobin is obtained from pure myoglobin by extracting out the heme group. Using
UV-visible spectrometry we are able to monitor complete removal of heme group. This
extraction initiates the disruption of myoglobin’s tertiary conformation. Apomyoglobin consists
of 8 α-helices labeled A through H. These helices unfold when the protein is subjected to pH
decrease. It is more compact at about pH 7 and unfolds as we change to more acidic
environment. At about pH 2 we have A, G, H core with the rest of the helices unfolded. At pH
values slightly lower than 2, A, G and H helices are also unfolded. When the protein is not
denatured, the refolding process is done by changing the pH towards neutral. At pH 2, G and H
helices refold and at pH 4 A-helix refolds as well. We have been able to label A-helix with
protons and the rest of the protein with deuterium using H2O and D2O respectively. We have
then used temperature change to initiate the unfolding of protonated A-helix. When dissolved in
D2O, the protonated A-helix at elevated temperatures exchanges its N-H protons to N-D
deuterons. Temperatures were elevated from 0˚ C to 80˚C and changes in Am III peak intensities
were observed using UV resonance Raman spectroscopy. These peak changes were quantified
18
and used to calculate the number of N-H bonds present between 0˚ C and 80˚C. The number of
N-H bonds decreased with increase in temperature indicating the unfolding of A-helix.
3.2 INTRODUCTION
Apomyoglobin (apoMb.) is obtained from myoglobin’s single polypeptide chain of
helical secondary structures (8 α-helices, labeled A-H) by extracting out the heme group. For
protein folding studies, it has been found that heme removal results in the formation of a partially
unfolded intermediate. This intermediate structure can be further destabilized with acid
denaturation. In recent work, Chi and Asher have used UV resonance Raman (UVRRS)
spectroscopy to examine how change in pH of the environment induces changes in secondary
structure of apoMb.34 Resonance Raman involves excitation within the region of the electronic
absorption of the peptide bond. In proteins, the amide region is excited by wavelengths 204 nm
and 206.5 nm. It is in this range of wavelengths that the amide bonds undergo resonance
enhancement in electronic transition from the ground to the excited state.
Using UV resonance Raman spectroscopy it has been possible to determine the existence
of three unfolding phases in apoMb (Figure 4).34 The native state of apoMb (pH 7) was shown to
be 62% α-helical. When the functional heme group is removed there is destabilization of the C,
D, and F helices (Figure 4, II). At pH 4, a stable AGH intermediate was found to exist, protecting
the Trp 14 residues, but with increased water exposure for the Trp 7 residue located on A-helix
(figure 4, III). Upon further decrease in pH, the protein further unfolds to only G and H helices
(Figure 4, IV).
19
Based on these studies by Chi and Asher, we have done further research to attempt to
label a single helix A and observe how it behaves at different pH values as shown in the next
sections of this chapter.
Figure 4. Model for acid denaturation of apoMb and holoMb. The helical structure is represented
as ribbons. Lines represent unfolded structures or the so-called β-strands. Adapted from Chi Z.
and Asher S.A.34
3.3 EXPERIMENTAL
Horse heart myoglobin was obtained from Sigma Aldrich and used as the only source of
apomyoglobin (apoMb). Extraction of the heme group was performed using 2-butanone as
described in Teale.35 Absorption spectra in the UV-Vis region were taken to confirm complete
removal of the heme group as shown in figure 5.
20
0
1
2
3
200 400 600 800
0
1
2
3
400 200 400 600 800
Abso
rban
ce (A
U)
Wavelength (nm)
Sore
t ban
d
Afte
r he
me
extrac
tion
Figure 5. Absorption spectra of myoglobin before and after extraction of heme group. The Soret
band at 400 nm disappears which indicates loss of the heme group. Concentration of the original
solution was 5 mg/ml.
The freshly made apoMb. was purified using a Waters Delta Prep HPLC with C-18 preparative
column. Two detectors at 278 nm and 220 nm were used to detect the protein. The solution
containing 50% H2O, 50% ACN and 0.1% TFA with 20µL of 1 M DTT (Dithiothereitol) was
prepared. ApoMb has two Tryptophan chains on the A-helix and were detected after 31 minutes
by the both detectors as shown in figure 6. The samples were collected at different times in
21
different bottles corresponding to different peaks and in order to verify which bottle contained
ApoMb, Mass Spectrometer (LC/MS) was used.
AU
Minutes
Figure 6. HPLC was used to purify apomyoglobin. The top spectrum is obtained by using 220
nm detector and bottom uses the 278 nm detector. The peak at 31st minute appears in both
spectra and corresponds to apomb.
The mass spectrum is shown in figure 7. Six different samples were injected directly into
the LC/MS. The apoMb sample was identified using a known apoMb sample. The peak was
analyzed as shown in figure 7 and table 1.
22
M/Z
Rel
ativ
e In
tens
ity
Figure 7. Mass spectrum of Apomyoglobin. The above m/z ratios are indicative of a multi-
charged molecule. The charges on this protein are as high as 29+ as shown in a table that
follows.
The LC/MS was a true indicator of apoMb presence in the purified sample. The m/z
ratios obtained correspond to those of apomyoglobin as calculated in the table 1 below. The
calculation uses the molecular weight of horse heart myoglobin.
23
Table 1. Calculated charges z obtained from the mass spectrum of apomyoglobin in figure 6. The
formula used for this calculation is shown on the table. The molecular weight (mw) of horse
heart myoglobin in this case is 16947 g.
The sample was then centrifuged to remove all the organic solvents (mostly acetonitrile).
It was then lyophilized for 12 hours to produce powdered apoMb. The new powder was kept at
room temperature until needed. For CD experiments, 10 – 20 µM solutions were prepared. For
Raman experiments, about 50 µM solution in D2O and 60 µM solution in H2O were prepared.
• To calculate these charges,( M/Z)
• m/z= (mw + z*mass) / z14 1211.50
• Where mw is molecular 15 1130.80weight, 16 1060.19
• m/z is mass to charge ratio17 997.88• z-number of charges18 942.50• - z* charge of one proton or
19 892.95 sodium depending on which 20 848.35 one is attached21 808.00
22 771.32
23 737.83
24 707.13
25 678.88
26 652.81
27 628.67
28 606.25
29 585.38
Z
24
3.4 RESULTS AND DISCUSSION
A Jasco 715 CD spectropolarimeter was used to study the behavior of apoMb at different
pH values at 25 ºC. The sample was dissolved in D2O and pH adjustments were made using DCl.
At pH 2 and lower there was no α-helix observed. We then saw α-helical conformations appear
as the pH increased towards neutral as seen in figure 8.
CD
(mde
g)
-25
-20
-15
-10
-5
0
5
10
180 190 200 210 220 230 240 250
Wavelength (nm)
1.7
2.0
4.0
5.7
6.3
PH
Figure 8. CD spectra of apoMb at different pH values from 1.74 to 6.30. It shows no helical
conformation at pH 2 and below. We see α-helix conformation appear at pH 4 and above.
Concentrations are about 20 µM ApoMb in D2O. Temperature was 25º C.
For UV resonance Raman experiments apoMb was incubated in D2O for 16 hrs at pH
1.74 to deuterate all the helices. ApoMb concentrations of 50 µM in D2O and 60 µM in H2O
25
were used. The goal was to study the behavior of the two solutions at pD 4 and pH 4. Figure 9
shows UV resonance Raman spectra at temperature 10 ºC.
1200 1300 1400 1500 1600 1700
Ram
an In
tens
ity /
AU
Raman Shift / cm-1-
A) In D2O
B) In H2O
C)H2O-D2O
Am II
I
Am I’Am
II’
Figure 9. UVRRS excited at 204 nm at 10º C: (A) ApoMb after 16hrs in D2O, pD 4.0.
(B)ApoMb in H2O, pH 4.0. (C) Difference spectra.
From the spectra, we see that the Am II’ band in D2O disappears in H2O. We also see the
Am III band in H2O solution disappears in D2O solution because of N-D deuteration. The Am III
band disappears upon N-D deuteration because decoupling between N-H bending and C (O)-N
stretching forms an Am II’ band which is almost pure C (O)-N stretching.
The fully deuterated sample (pD <2, incubated in D2O) was brought to pD 2.05 where the
G and H helices refold. It was then frozen in liquid nitrogen and lyophilized. At this point the A-
helix, which is still unfolded was protonated by dissolving the powder in H2O at pH 3.5. The
solution was raised to pH 4 where A-helix refolds with all its N-H protonated. It was frozen and
26
lyophilized to obtain the solid form with a protonated A-helix. Upon dissolution in D2O, this
sample(A-H) initially contains only protonated peptide bonds in the A-helix.
The solid apoMb with a protonated A-helix was dissolved in D2O at pD 4.08. This
allowed deuterium exchange on the A-helix. The pD 4.08 value was chosen because exchange is
slow enough at this pD to prevent complete deprotonation before Raman experiments are
performed.37 Exchange happens due to unfolding of A-helix exposing its N-H bond to D2O
environment. Figure 10. shows 204 nm resonance Raman spectra taken at different temperatures.
Am
II’
1000 1200 1400 1600 1800 1000 1200 1400 1600 1800
Ram
an In
tens
ity /
au
Raman Shift / cm-1
0 C 10 C 30 C 60 C80 C
PH 4.08
Am
I
Am
III
Figure 10. 204 nm UV resonance Raman spectra of apoMb sample measured between 0 and 80
ºC. Three 5 min spectra were recorded and summed. They were normalized with respect to the
Am I’ band. The change in AmIII region is due to peptide bond deuteration which is facilitated
by A-helix melting.
27
The values of these Fig.10 peak intensities are given in table 2 . The Am III peak
vanishes. As the temperature increases the A-helix unfolds and some N-H bonds are exchanged
to N-D due to presence of a D2O environment.
0ºC10ºC30ºC60ºC80ºC
Ram
an In
tens
ity A
U
1300 1600Raman Shift /cm-1
ΔAmIII ΔAmI’
Figure 11. Raman spectra used to estimate peak intensities IAmI’ and IAmIII in the Am I’ and Am
III regions, respectively. Am I’ peaks are used to normalize the spectra. The peak intensity
results are shown in table 2 below.
28
Temperature [deg
C] AmI' Intensity /mm AmIII Intensity /mm
No. of N-H bonds in Am
III region
0 20 6.2 62
10 20 6.0 61
30 20 4.5 46
60 20 4.2 43
80 20 4.0 41
Table 2. Table shows Amide I’, and III peak heights obtained from figure 11. These intensities
IAmI’ and IAm III, were obtained by normalizing all spectra with respect to Am I’ band and using a
ruler to measure them as shown in figure 11 above. The last column is obtained using equation
26 (also explained in figure 12).
We can calculate the number of peptide bonds contributing to the Am III region.
nAm III =(I Am III nAmI’’ σ AmI’)/ (I Am I’ σ AmIII) (26)
Where, I Am III and I Am I’ are the integrated intensities of the Am III and Am I’ bands,
respectively. The number of peptide bonds per protein contributing to the Am III and Am I’
band intensities are nAmIII and nAmI’, respectively, whereas the σ AmI’ and σ AmIII are cross sections
of the Am I’ and Am III bands, respectively.
Assuming equal Am I and Am I’ cross sections, we can use the cross sections of the Am I
and Am III bands of the α-helical peptides determined by Chi et al.(39) to determine the number
of peptide bonds in α-helix like conformations.
29
No. of peptide bonds in the Am III region vs Temperature
20
30
40
50
60
70
80
0 20 40 60 80 100
Temperature [deg C]
No.
of N
-H b
onds
8
7
6
5
4
3
Am
III R
aman
Inte
nsity
[mm
]
Figure 12. This graph is plotted using data from table 2 which were generated from equation 26.
IAmI’ and IAmIII were measured in figure 11 and shown in table 2. The number of N-H bonds
contributing to Am III band is nAmIII and to Am I’ band is nAmI’, =152 peptide bonds/protein. The
α-helix cross sections are σ AmI’ =32 mbarn molecule-1 sr-1 and σ AmIII =24 mbarn molecule-1 sr-1,
obtained from Chi et al.(39). Also superimposed is the dependence on temperature of Am III
intensity normalized with respect to Am I’.
We show in the above figure, that, the number of peptide bonds decreases by
approximately 20 peptide bonds between temperatures 0ºC and 80ºC. The observed initial
number of peptide bonds is a higher value than the twenty expected from a fully protonated A-
helix i.e. 60 peptide bonds at 0 ºC. This could be a result of tryptophan’s aromatic band
contributing to the AmIII region hence resulting to more intensity than expected. The
contribution of tryptophan aromatic band in Am III region is evident from Trpcage study by
30
Ahmed et al. (1). The loss in number of peptide bonds is approximately twenty which is same
number of peptide bonds in the A-helix. We can therefore conclude that this is an evidence of A-
helix unfolding.
3.5 CONCLUSION
Preliminary work on this project involved careful extraction and purification of
apomyoglobin samples. The methods used included fractional extraction of heme group from
horse heart myoglobin using MEK-water mixture and HPLC purification. UV-Vis spectrometry
was used to confirm extraction of heme group as seen in figure 5. LCMS was then used to
identify apomyoglobin from the purification process. CD spectroctrometry was also used to
identify some helical confirmation of apomyoglobin in different pH environment.
The UV resonance Raman experiments were run to study the conformation of A-helix.
Experimental results show a decreasing intensity in Am III region as seen in figure 12. We also
observed a decreasing number of N-H bonds with increasing temperature same figure. The
changes are due to exchange of protonated A-helix with deuterium. It is therefore evident from
the above amide band changes that the A-helix can be labeled and can unfold with increase in
temperature.
31
4.0 FUTURE WORK
4.1 TIME RESOLVED MEASUREMENTS OF APOMYOGLOBIN UNFOLDING
T-jump experiment can be done to study the equilibrium conformation of apomyoglobin.
The T-jump experimental set-up is described below.
Figure 13. T-jump spectrometer consisting of an Nd: YAG laser, 2 H2 Raman shifters (R1 and
R2), a thermostated flow cell sample circulator (FC) and an ICCD detector (SP).reproduced from
Lednev I. K. reference38
32
There are 2 beams involved in a T-jump experiment, the pump and the probe.38 The heating
pump is acquired by Raman shifting the YAG fundamental in H2 to 1.9 µm i.e. 1st stokes. The
probe beam is obtained by shifting the YAG 3rd harmonic to 204 nm (5th anti-stokes). The time
delay between the pump and probe is computer controlled.
The sample is circulated though the two laser beams. The IR heating pump beam is used
to heat the solvent (i.e. create a temperature-jump) and the probe is used to detect changes in the
protein that the pump beam initiated. We can then take spectra within different time scales (i.e. 3
ns, 5 ns, 10 ns etc) and observe unfolding process.
In future work we would like to do T-jump to study the unfolding kinetics of
apomyoglobin. We will prepare apomyoglobin with a labeled A-helix as explained in chapter 3
of this book. We will then initiate unfolding by performing a rapid T-jump with IR pulse as
explained in the experimental setup above. A 30 ºC T-jump can be obtained from the sample
initially at 35 ºC. This will initiate the A-helix melting. The UVRRS can then be used to examine
the unfolding dynamics of protonated A-helix which would show a decreased AmIII integrated
peak areas. A concurrent increase of AmII’ would be observed. After successfully studying the
A-helix, we will use the same methodology on G and H helices.
33
REFERENCES
1. Ahmed, Z. UV Resonance Raman spectroscopy of Proteins. Comprehensive thesis
University of Pittsburgh, Pittsburgh, 2003.
2. Frank, H. S., Free Volume and Entropy in Condensed Systems I. General Principles.
Fluctuation Entropy and Free Colume in Some Monatomic Crystals. Journal of Chemcial
Physics 1945, 13, 478-492.
3. Frank, H. S., Free Volume and Entropy in Condensed Systems II. Entropy of
Vaporization in Liquids and the Pictorial Theory of the Liquid State. Journal of Chemcial
Physics 1945, 13, 493-507.
4. Frank, H. S.; Evans, M. W., Free Volume and Entropy in Condensed Systems III.
Entropy in Binary Liquid Mixtures; Partial Molal Entropy in Dilute Solutions; Structure
and Thermodynamics in Aqueous Electrolytes. Journal of Chemcial Physics 1945, 13,
507-532.
5. Asher, S. A.; Chi, Z., UV resonance Raman determination of a-helix melting during the
acid denaturation of myoglobin. NATO ASI Series, Series E: Applied Sciences 1997, 342,
(Biomolecular Structure and Dynamics), 263-270.
6. Asher, S. A.; Chi, Z.; Li, P., Resonance Raman examination of the two lowest amide
pipi* excited states. Journal of Raman Spectroscopy 1998, 29, (10/11), 927-931.
34
7. Asher, S. A.; Ianoul, A.; Mix, G.; Boyden, M. N.; Karnoup, A.; Diem, M.; Schweitzer-
Stenner, R., Dihedral ψ angle dependence of the amide III vibration: A uniquely sensitive
UV resonance Raman secondary structural probe. J. Am. Chem. Soc. 2001, 123, (47),
11775-11781.
8. Asher, S. A.; Jones, C. M., UV resonance Raman spectroscopy: a new technique for
speciation of aromatics in complex matrixes. Preprints of Papers - American Chemical
Society, Division of Fuel Chemistry 1986, 31, (1), 170-80.
9. Asher, S. A.; Ludwig, M.; Johnson, C. R., UV Resonance Raman excitation profiles of
the aromatic amino acids. Journal of the American Chemical Society 1986, 108, (12),
3186-97.
10. Asher, S. A.; Mikhonin, A. V.; Bykov, S. V., UV Raman Demonstrates that a-helical
Polyalanine Peptides Melt to Polyproline II Conformations. J. Am. Chem. Soc. 2004, 126,
(27), 8433-8440.
11. Asher, S. A.; Murtaugh, J. L., UV Raman excitation profiles of imidazole, imidazolium,
and water. Applied Spectroscopy 1988, 42, (1), 83-90.
12. Austin, J. Raman spectroscopic studies of selected enzyme systems. 1989.
13. Boyden, M. N.; Asher, S. A., UV Raman Studies of Peptide Conformation Demonstrate
That Betanova Does Not Cooperatively Unfold. Biochemistry 2001, 40, (45), 13723-
13727.
14. Clore, G. M.; Gronenborn, A. M., NMR of Proteins. 1993.
15. Jas, G. S.; Eaton, W. A.; Hofrichter, J., J. Phys. Chem. B, 2001, 105,261-272.
35
16. Manas, E. S.; Getahun, Z.; Wright, W. W.; DeGrado, W. F.; Vanderkooi, J. M., Infrared
Spectra of Amide Groups in a-Helical Proteins: Evidence for Hydrogen Bonding between
Helices and Water. J. Am. Chem. Soc. 2000, 122, (41), 9883-9890.
17. Ozdemir, A.; Lednev, I. K.; Asher, S. A., Comparison between UV Raman and Circular
Dichroism Detection of Short a Helices in Bombolitin III. Biochemistry 2002, 41, (6),
1893-1896.
18. Kaltashov; A., I.; Eyles; J, S., Mass Spectrometry Reviews 2002, 21, 1893-1896.
19. Pain, R. H., Mechanisms of Protein Folding. ed.; Oxford University Press: New York,
NY, 1994; 'Vol.' p.
20. Ahmed, Z.; Illir, A. B.; Mikhonin, A. V.; Asher, S. A., UV-Resonance Raman Thermal
Unfolding Study of Trp-Cage Shows That It Is Not a Simple Two-State Miniprotein. J.
Am. Chem. soc. 2005, 127, 10943 -10950.
21. Neidigh, J. W.; Fesinmeyer, R. M.; Andersen, N. H., Designing a 20-residue protein.
Nature Structural Biology 2002, 9, (6), 425-430.
22. Boyden, M. N. Spectroscopic Investigation of Peptides. University of Pittsburgh,
Pittsburgh, 1998.
23. Asher, S. A., Resonance Raman spectroscopy of hemoglobin. Methods in Enzymology
1981, 76, (Hemoglobins), 371-413.
24. Asher, S. A., UV Raman Spectroscopy for Analytical, Physical, and Biophysical
Chemistry. Anal. Chem. 1993, 65, 59A-66A, 201A-210A.
25. Myers, A. B., Chem. Rev. 96, 3, 911-926.
26. Lednev, I. K.; Karnoup, A. S.; Sparrow, M. C.; Asher, S. A., Nanosecond UV resonance
Raman examination of initial steps in a-helix secondary structure evolution. Spectroscopy
36
of Biological Molecules: New Directions, European Conference on the Spectroscopy of
Biological Molecules, 8th, Enschede, Netherlands, Aug. 29-Sept. 2, 1999 1999, 11-12.
27. Creamer, T. P.; Srinivasan, R.; Rose, G. D., Modeling unfolded states of peptides and
proteins. Biochemistry 1995, 34, (50), 16245-50.
28. Creamer, T. P.; Srinivasan, R.; Rose, G. D., Modeling Unfolded States of Proteins and
Peptides. II. Backbone Solvent Accessibility. Biochemistry 1997, 36, (10), 2832-2835.
29. Shi, Z.; Olson, C. A.; Rose, G. D.; Baldwin, R. L.; Kallenbach, N. R., Polyproline II
structure in a sequence of seven alanine residues. Proceedings of the National Academy
of Sciences of the United States of America 2002, 99, 9190-9195.
30. Kelly, M. A.; Chellgren, B. W.; Rucker, A. L.; Troutman, J. M.; Fried, M. G.; Miller, A.-
F.; Creamer, T. P., Host-Guest Study of Left-Handed Polyproline II Helix Formation.
Biochemistry 2001, 40, 14376-14383.
31. Long, D. A., Raman Spectroscopy. 1977, 3-84.
32. Asher, S. A., Ultraviolet Raman Spectroscopy. Handbook of Vibrational Spectroscopy,
Vol 1, 2001, John Whiley & Sons, Ltd.
33. Asher, S. A., Analytical Chemistry 1993, 65(2), 59A-66A.
34. Chi, Z.; Asher, S. A., Ultraviolet resonance Raman examination of horse apomyoglobin
acid unfolding intermediates. Biochemistry 1999, 38, (26), 8196-8203.
35. Teale, F. W. J., Cleavage of the hemoprotein link by acid methyl ethyl ketone. Biochim.
Biophys. Acta 1959, 35, 543.
36. Asher, S.; Sauer, K., Resonance Raman spectroscopy of manganese(III) etioporphyrin I
at the pi -> pi* and charge transfer bands: The use of charge transfer bands to monitor the
37
38
complexation state of metalloporphyrins. Journal of Chemical Physics 1976, 64, (10),
4115-25.
37. Englander, S. W., Protein folding intermediates and pathways studied by hydrogen
exchange. Annu. Rev. Biophys. Biomol. Struct. 2000, 29, 213-238, 3 Plates.
38. Lednev, I. K.; Karnoup, A. S.; Sparrow, M. C.; Asher, S. A., a-Helix peptide folding and
unfolding activation barriers: A nanosecond UV resonance Raman study. J. Am. Chem.
Soc. 1999, 121, (35), 8074-8086.
39. Chi, Z.; Chen, X. G.; Holtz, J. S. W.; Asher, s. A.; UV Resonance Raman-Selective
Amide Vibrational Enhancement: Quantitative Methodology for Determining Protein
Secondary Structure. Biochemistry 1998, 37, 2854-2864